Applications in Neural and Symbolic Artificial Intelligence

Dear Colleagues,

Neural-based AI, often referred to as deep learning, leverages artificial neural networks (ANNs) to model complex patterns in data. These networks, inspired by the human brain's structure, consist of interconnected neurons that process inputs and learn from vast amounts of data. Deep learning has revolutionized various fields, including image and speech recognition, natural language processing, and autonomous systems. However, neural networks require extensive data for training and often operate as "black boxes", lacking interpretability and the ability to reason logically.

Symbolic AI, also known as classical AI, relies on the manipulation of symbols and rules to represent knowledge and perform reasoning. This approach excels in domains requiring logical inference, problem-solving, and the use of expert knowledge. Symbolic AI systems can explain their reasoning processes and make decisions based on predefined rules and knowledge bases. However, they struggle with learning from raw data and adapting to new, unforeseen situations, making them less flexible compared to neural networks.

Neuro-symbolic AI aims to integrate the learning capabilities of neural networks with the reasoning and knowledge representation strengths of symbolic AI. This hybrid approach addresses the limitations of both methods by enabling systems to learn from data while also reasoning with structured knowledge. Neuro-symbolic AI can enhance interpretability, allowing AI systems to provide explanations for their decisions and improve their ability to generalize from limited data. Applications of neuro-symbolic AI span various fields, including natural language processing, robotics, cybersecurity, smart cities, education, and various fields of science and technology.

This Special Issue seeks high-quality, original research articles, reviews, and case studies that delve into the innovative use of neural and symbolic artificial intelligence in various aspects of science and technology. Topics of interest include, but are not limited to, the following:

• Advancements in deep learning techniques for neuro-symbolic integration: exploring innovations in neural network architectures and training algorithms that facilitate the integration of symbolic reasoning capabilities to enhance the synergy between neural and symbolic components.
• Innovative approaches to knowledge representation in AI: developing new methods for encoding knowledge that can be utilized by both neural and symbolic systems, including techniques for integrating knowledge graphs with neural networks, including large language models (LLMs), to enhance reasoning.
• Cognitive science perspectives on neuro-symbolic AI: applying insights from cognitive science about how humans integrate perceptual and symbolic information to develop more human-like AI systems.
• Enhancements in machine learning algorithms through symbolic methods: utilizing symbolic methods to improve machine learning algorithms' accuracy, efficiency, and interpretability, supported by case studies demonstrating the benefits of this hybrid approach in various applications.
• Case studies and applications of neuro-symbolic AI in science, industry and academia: showcasing detailed case studies of neuro-symbolic AI implementation in real-world scenarios, and analyzing the impact and benefits of these applications in different sectors.
• The role of formal logic in enhancing AI systems: exploring how formal logic can improve the robustness and reliability of AI systems, and methods for integrating formal logic with neural networks to enhance decision-making processes.
• Ethical considerations and trustworthiness in neuro-symbolic AI: examining the ethical implications of neuro-symbolic AI systems and strategies for ensuring transparency, fairness, and accountability in these systems (SpringerLink).
• Future directions and the potential of human-like interaction in AI: exploring future trends and developments in neuro-symbolic AI, and potential applications in creating more natural and intuitive human–AI interactions.

Dr. Bikram Bhuyan
Dr. Manolo Dulva Hina
Prof. Dr. Amar Ramdane-Cherif
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• machine learning
• artificial neural networks
• deep learning
• symbolic computing
• knowledge representation
• formal logic
• trustworthy AI
• large language models